Commit
•
6ca1a0e
1
Parent(s):
1d7af98
initial release
Browse files- README.md +44 -0
- config.json +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +9 -0
- supar.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +16 -0
- vocab.txt +0 -0
README.md
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- "ko"
|
4 |
+
tags:
|
5 |
+
- "korean"
|
6 |
+
- "token-classification"
|
7 |
+
- "pos"
|
8 |
+
- "dependency-parsing"
|
9 |
+
datasets:
|
10 |
+
- "universal_dependencies"
|
11 |
+
license: "cc-by-sa-4.0"
|
12 |
+
pipeline_tag: "token-classification"
|
13 |
+
widget:
|
14 |
+
- text: "홍시 맛이 나서 홍시라 생각한다."
|
15 |
+
---
|
16 |
+
|
17 |
+
# roberta-large-korean-morph-upos
|
18 |
+
|
19 |
+
## Model Description
|
20 |
+
|
21 |
+
This is a RoBERTa model for POS-tagging and dependency-parsing, derived from [klue/roberta-large](https://huggingface.co/klue/roberta-large) and [morphUD-korean](https://github.com/jungyeul/morphUD-korean). Every morpheme (형태소) is tagged by [UPOS](https://universaldependencies.org/u/pos/)(Universal Part-Of-Speech).
|
22 |
+
|
23 |
+
## How to Use
|
24 |
+
|
25 |
+
```py
|
26 |
+
from transformers import AutoTokenizer,AutoModelForTokenClassification,TokenClassificationPipeline
|
27 |
+
tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/roberta-large-korean-morph-upos")
|
28 |
+
model=AutoModelForTokenClassification.from_pretrained("KoichiYasuoka/roberta-large-korean-morph-upos")
|
29 |
+
pipeline=TokenClassificationPipeline(tokenizer=tokenizer,model=model,aggregation_strategy="simple")
|
30 |
+
nlp=lambda x:[(x[t["start"]:t["end"]],t["entity_group"]) for t in pipeline(x)]
|
31 |
+
print(nlp("홍시 맛이 나서 홍시라 생각한다."))
|
32 |
+
```
|
33 |
+
|
34 |
+
or
|
35 |
+
|
36 |
+
```py
|
37 |
+
import esupar
|
38 |
+
nlp=esupar.load("KoichiYasuoka/roberta-large-korean-morph-upos")
|
39 |
+
print(nlp("홍시 맛이 나서 홍시라 생각한다."))
|
40 |
+
```
|
41 |
+
|
42 |
+
## See Also
|
43 |
+
|
44 |
+
[esupar](https://github.com/KoichiYasuoka/esupar): Tokenizer POS-tagger and Dependency-parser with BERT/RoBERTa/DeBERTa models
|
config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:eebadebe77ac215e7e3d092db1d4d302f28d15495bb423aeaa0805881094f601
|
3 |
+
size 1346220849
|
special_tokens_map.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "[CLS]",
|
3 |
+
"cls_token": "[CLS]",
|
4 |
+
"eos_token": "[SEP]",
|
5 |
+
"mask_token": "[MASK]",
|
6 |
+
"pad_token": "[PAD]",
|
7 |
+
"sep_token": "[SEP]",
|
8 |
+
"unk_token": "[UNK]"
|
9 |
+
}
|
supar.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b6709f1bbce1c7babe3c9e5adad4b36aaec899028f20a25e0210caf94bf229d6
|
3 |
+
size 1394978533
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "[CLS]",
|
3 |
+
"cls_token": "[CLS]",
|
4 |
+
"do_basic_tokenize": true,
|
5 |
+
"do_lower_case": false,
|
6 |
+
"eos_token": "[SEP]",
|
7 |
+
"mask_token": "[MASK]",
|
8 |
+
"model_max_length": 512,
|
9 |
+
"never_split": null,
|
10 |
+
"pad_token": "[PAD]",
|
11 |
+
"sep_token": "[SEP]",
|
12 |
+
"strip_accents": null,
|
13 |
+
"tokenize_chinese_chars": true,
|
14 |
+
"tokenizer_class": "BertTokenizerFast",
|
15 |
+
"unk_token": "[UNK]"
|
16 |
+
}
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|